1. Technical Field
The present principles relate generally to power management systems and, in particular, to geospatial optimization for resilient power management equipment.
2. Description of the Related Art
Presently, in server environments utilizing backup power supplies (e.g., uninterruptable power supplies (UPS) or small on-site generators), standard practice is to assume the worst. That is, power will not be restored before the backup power supply is exhausted. Under this assumption, the current practice to avoid an uncontrolled shut down is to power down all systems immediately, which is executed with varying degrees of elegance. Backup power supplies are also selected under this assumed mode of operation in the event of a power outage.
There are a number of limitations with this approach. First, if the systems were providing some service, then time spent offline equates to a decrease in the quality of service and, potentially, compromised revenue (e.g., with respect to web services, cloud computing providers, and so forth). Even in the case where the service has little value during a power failure (since users of the service may also be experiencing power failure), the time required to restore the service can contribute significantly to the duration of the outage. For example, a 15 second power outage could result in 5 minutes of service outage. The second limitation arises from the increased likelihood of device failure during power on. For example, hard drives and other system devices have an increased likelihood of device failure during power on. Thus, in large server environments, each power on can contribute substantially to device failures.